Our previous work has determined that fibronectin (FN) is released from the cell surface by tumor promoters, a release inhibited by treatment with retinoids. We were able to conclude that this release is a necessary (but not sufficient) step in tumor promotion. We found that the TPA-released FN is phosphorylated less than FN from untreated cells. The proposed work will investigate: (I) whether the affinity of the underphosphorylated FN from TPA-treated cells is lower, both in binding to the matrix assembly receptor (pool I) and the matrix itself (pool II), from medium, cell-surface and endogenously-derived FN; (II) whether protein phosphokinases A and C are involved in FN phosphorylation and its inhibition by TPA (and activation by retinoids), using kinase activation and kinase inhibitors; (III) whether the number of location of phosphate groups changes in FN as a result of TPA treatment, and whether the amino terminal 70,000 mol. wt. proteolytic fragment of FN which was found by Mosher et al (1) to bind to the matrix assembly receptor, has phosphate groups which may be involved in binding; whether the self-assembly in matrix fiber formation is affected by TPA and by phosphorylation. These experiments will be done first with untransformed human lung fibroblasts and later with a line (JB-6) of initiated (by not promoted) mouse epidermal cells.